Diagnostic and/or prognostic biomarkers for cholangiocarcinoma (CCA) are still insufficient with poor prognosis of patients. To discover a new CCA biomarker, we constructed our secretome database of three CCA cell lines and one control cholangiocyte cell line using GeLC-MS/MS. We selected candidate proteins by five bioinformatics tools for secretome analysis. The inclusion criteria were as follows: having predicted signal peptide or being predicted as non-classically secreted protein; together with having no transmembrane helix and being previously detected in plasma and having the highest number of signal peptide cleavage sites. Eventually, apurinic/apyrimidinic endodeoxyribonuclease 1 (APEX1) was selected for further analysis. To validate APEX1 as a bio-marker for CCA, serum APEX1 levels of 80, 39, and 40 samples collected from CCA, benign biliary diseases (BBD), and healthy control groups, respectively, were measured using dot blot analysis. The results showed that serum APEX1 level in CCA group was significantly higher than that in BBD or healthy control group. Among CCA patients, serum APEX1 level was significantly higher in patients having metastasis than in those without metastasis. The higher level of serum APEX1 was correlated with the shorter survival time of the patients. Serum APEX1 level might be a diagnostic and prognostic biomarker for CCA.
Conditional random fields for the prediction of signal peptide cleavage sites